Applications of Artificial Intelligence in Climate-Resilient Smart-Crop Breeding

Muhammad Hafeez Ullah Khan, Shoudong Wang, Jun Wang, Sunny Ahmar, Sumbul Saeed, Shahid Ullah Khan, Xiaogang Xu, Hongyang Chen, Javaid Akhter Bhat, Xianzhong FengInternational Journal of Molecular Sciences 2022 · DOI

TL;DR

Artificial Intelligence (AI) is revolutionizing crop breeding by enabling faster, more accurate analysis of massive "omics" datasets. This review highlights how AI tackles challenges in high-throughput phenotyping, gene function analysis, and the crucial genotype-phenotype link, paving the way for climate-resilient smart crops.

What Problem Does This Paper Address?

Modern agriculture faces the dual challenge of feeding a growing global population while adapting to a changing climate. Developing climate-resilient crops is paramount, but traditional breeding methods are slow and struggle to keep pace with the complexity of genetic and environmental interactions. The sheer volume and intricacy of data generated by "omics" technologies (genomics, phenomics, enviromics) present a significant bottleneck. Specifically, linking an organism's genetic makeup (genotype) to its observable traits (phenotype) under diverse environmental conditions (envirotyping) is a major hurdle. This paper addresses how AI can overcome these limitations to accelerate the development of improved crop varieties.

Methods

This review focuses on the application of AI, particularly machine learning (ML) and deep learning (DL), across various aspects of crop breeding. Key AI techniques discussed include:

  • Computer Vision: For analyzing image data in high-throughput phenotyping.
  • Machine Learning Algorithms: Such as k-Nearest Neighbors (kNN), Random Forest (RF), and Support Vector Machines (SVM) for prediction and classification tasks.
  • Deep Learning: For more complex pattern recognition and data analysis, especially in gene function prediction and image segmentation.
  • AI-driven Data Integration: Combining diverse "omics" data (genomics, phenomics, enviromics) to build comprehensive models.

The paper emphasizes AI's role in automating data collection, improving analytical accuracy, and facilitating the integration of disparate data sources.

Datasets

The paper discusses the use of various "omics" datasets, including:

  • Genomic Data: Including genome sequences and SNP (Single Nucleotide Polymorphism) data.
  • Phenotypic Data: Collected through high-throughput phenotyping, both in controlled (indoor) and field (outdoor) environments.
  • Envirotypic Data: Information on environmental conditions and their impact on crop performance.
  • Crop Germplasm and Mapping Populations: Used for generating large-scale datasets for analysis.

The paper does not introduce new datasets or benchmarks but rather reviews how existing and emerging AI techniques can be applied to these data types.

Key Insights and Ideas

  • AI as a Bridge: AI acts as a crucial bridge, translating massive, complex "omics" data into actionable insights for breeders.
  • Democratizing Data Analysis: AI tools, especially when integrated with cloud-based platforms and mobile devices, can make advanced analysis more accessible to farmers and breeders.
  • Beyond Prediction: While AI excels at prediction (e.g., yield, trait performance), its true power lies in its ability to uncover complex, non-obvious relationships within biological systems.
  • Envirotyping's Potential: Envirotyping, combined with AI, is poised to become a critical third "typing" technology, enabling a deeper understanding of genotype-by-environment interactions.
  • Accelerating Discovery: AI can significantly speed up gene identification and functional analysis, which are critical bottlenecks in traditional breeding.

Key Results

The paper highlights several successful applications of AI in crop breeding:

  • Improved Phenotyping Accuracy: AI-assisted systems have shown success in identifying plant growth stages, segmenting crops and weeds, and phenotyping disease resistance. For example, ML algorithms achieved high accuracy (R² = 0.67) in predicting root yield in cassava.
  • Enhanced Gene Function Prediction: AI models are being developed to predict gene functions, regulatory elements, and the impact of genetic variations.
  • Genomic Selection and Prediction: ML methodologies are increasingly used for genomic prediction and selection, leading to more efficient breeding.
  • Integration of Multi-omics Data: AI facilitates the integration of genomics, phenomics, and enviromics data, leading to more comprehensive breeding strategies.

Limitations and Future Work

  • Data Accessibility: While AI thrives on big data, the availability of comprehensive, standardized plant "omics" datasets remains a challenge.
  • Computational Demands: Training sophisticated AI models, especially deep learning, requires significant computational resources.
  • Interpretability: Understanding why an AI model makes a certain prediction (explainable AI) is crucial for building trust and further scientific discovery, and this is an ongoing area of research.
  • Cost and Complexity: While AI promises efficiency, the initial investment in technology and expertise can be substantial.

Future work will likely focus on developing more robust and interpretable AI models, creating standardized data platforms, and further integrating AI into the entire crop breeding pipeline, from discovery to deployment. The ultimate goal is to leverage AI to create a new generation of climate-resilient, high-performing crops that ensure global food security.

Summary updated

Abstract

Recently, Artificial intelligence (AI) has emerged as a revolutionary field, providing a great opportunity in shaping modern crop breeding, and is extensively used indoors for plant science. Advances in crop phenomics, enviromics, together with the other "omics" approaches are paving ways for elucidating the detailed complex biological mechanisms that motivate crop functions in response to environmental trepidations. These "omics" approaches have provided plant researchers with precise tools to evaluate the important agronomic traits for larger-sized germplasm at a reduced time interval in the early growth stages. However, the big data and the complex relationships within impede the understanding of the complex mechanisms behind genes driving the agronomic-trait formations. AI brings huge computational power and many new tools and strategies for future breeding. The present review will encompass how applications of AI technology, utilized for current breeding practice, assist to solve the problem in high-throughput phenotyping and gene functional analysis, and how advances in AI technologies bring new opportunities for future breeding, to make envirotyping data widely utilized in breeding. Furthermore, in the current breeding methods, linking genotype to phenotype remains a massive challenge and impedes the optimal application of high-throughput field phenotyping, genomics, and enviromics. In this review, we elaborate on how AI will be the preferred tool to increase the accuracy in high-throughput crop phenotyping, genotyping, and envirotyping data; moreover, we explore the developing approaches and challenges for multiomics big computing data integration. Therefore, the integration of AI with "omics" tools can allow rapid gene identification and eventually accelerate crop-improvement programs.
BibTeX
@article{khan2022applications,
  author = {Muhammad Hafeez Ullah Khan and Shoudong Wang and Jun Wang and Sunny Ahmar and Sumbul Saeed and Shahid Ullah Khan and Xiaogang Xu and Hongyang Chen and Javaid Akhter Bhat and Xianzhong Feng},
  title = {{Applications of Artificial Intelligence in Climate-Resilient Smart-Crop Breeding}},
  journal = {International Journal of Molecular Sciences},
  year = {2022},
  doi = {10.3390/ijms231911156},
}
Figure